Oversampling techniques are widely used to rebalance imbalanced datasets. However, most of the oversampling methods may introduce noise and fuzzy boundaries for dataset classification, leading to the overfitting phenomenon. To solve this problem, we propose a new method (FSDR-SMOTE) based on Random-SMOTE and Feature Standard Deviation for rebalancing imbalanced datasets. The method first removes noisy samples based on the Tukey criterion and then calculates the feature standard deviation reflecting the degree of data discretization to detect the sample location, and classifies the samples into boundary samples and safety samples. Secondly, the K-means clustering algorithm is employed to partition the minority class samples into several sub-clusters. Within each sub-cluster, new samples are generated based on random samples, boundary samples, and the corresponding sub-cluster center. The experimental results show that the average evaluation value obtained by FSDR-SMOTE is 93.31% (93.16%, and 86.53%) in terms of the F-measure (G-mean, and MCC) on the 20 benchmark datasets selected from the UCI machine learning library.

Traffic forecasting plays an important role in intelligent transportation systems. However, the prediction task is highly challenging due to the mixture of global and local spatiotemporal dependencies involved in traffic data. Existing graph neural networks (GNNs) typically capture spatial dependencies with the predefined or learnable static graph structure, ignoring the hidden dynamic patterns in traffic networks. Meanwhile, most recurrent neural networks (RNNs) or convolutional neural networks (CNNs) cannot effectively capture temporal correlations, especially for long-term temporal dependencies. In this paper, we propose a spatial–temporal attention graph convolution network (STAGCN), which acquires a static graph and a dynamic graph from data without any prior knowledge. The static graph aims to model global space adaptability, and the dynamic graph is designed to capture local dynamics in the traffic network. A gated temporal attention module is further introduced for long-term temporal dependencies, where a causal-trend attention mechanism is proposed to increase the awareness of causality and local trends in time series. Extensive experiments on four real-world traffic flow datasets demonstrate that STAGCN achieves an outstanding prediction accuracy improvement over existing solutions.

In this paper, we studied the optomechanically induced transparency (OMIT) in a cavity optomechanical system containing a cubic nonlinear oscillator. In our system, a partially transparent, dielectric membrane was placed in the middle of the F-P cavity. Due to the partial transmission and reflective property of the membrane, the membrane was combined with both the mirrors on the left and right sides to form two cavities. When the system was driven by two coupling fields, we calculated the quantum fluctuation of the optomechanical system operators and showed the response of the cavity optomechanical system to the probe field. We found that the cubic nonlinearity led to a shift of the OMIT window, which moved towards a frequency less than the resonance frequency, and the absorption peak became significantly asymmetrical when OMIT appeared. The shift of the OMIT dip provided a method to detect the nonlinear effects of the system due to the existence of cubic anharmonic potential.

Quantum entanglement will play an important role in future quantum technologies. Here, we theoretically study the steady-state entanglement between a cavity field and a macroscopic rotating mirror in a Laguerre–Gaussian-(LG)-cavity optomechanical system with cross-Kerr nonlinearity. Logarithmic negativity is used to quantify the steady-state entanglement between the cavity and mechanical modes. We analyze the impacts of the cross-Kerr coupling strength, the cavity detuning, the input laser power, the topological charge of the LG-cavity mode, and the temperature of the environment on the steady-state optomechanical entanglement. We find that cross-Kerr nonlinearity can significantly enhance steady-state optomechanical entanglement and make steady-state optomechanical entanglement more robust against the temperature of the thermal environment.